1. Gaussian MRF Rotation-Invariant Features for Image Classification.
- Author
-
Deng, Huawu and Clausi, David A.
- Subjects
- *
PROBABILITY theory , *STOCHASTIC processes , *MARKOV processes , *MATHEMATICAL statistics , *ESTIMATION theory , *GAUSSIAN processes - Abstract
Features based on Markov random field (MRF) models are sensitive to texture rotation. This paper develops an anisotropic circular Gaussian MRF (ACGMRF) model for retrieving rotation-invariant texture features. To overcome the singularity problem of the least squares estimate method, an approximate east squares estimate method is designed and implemented. Rotation-invariant features are obtained from the ACGMRF model parameters using the discrete Fourier transform. The ACGMRF model is demonstrated to be a statistical improvement over three published methods. The three methods include a Laplacian pyramid, an isotropic circular GMRF (ICGMRF), and gray level cooccurrence probability features. [ABSTRACT FROM AUTHOR]
- Published
- 2004
- Full Text
- View/download PDF